Chronic pelvic pain significantly affects the health of up to 10 percent of women with endometriosis (Stratton, Fertil Steril 2006;86: 1302). We have recently published the results of a randomized, prospective, placebo-controlled trial of raloxifene (180 mg daily) used by women with chronic pelvic pain and endometriosis. This study was one of the largest randomized studies of medical therapy for endometriosis and, unlike other studies of endometriosis and pain, adhered to stringent entry criteria, including only those with biopsy-proven disease. Unexpectedly, women treated with the selective estrogen-receptor modulator raloxifene experienced return of chronic pelvic pain sooner than those treated with placebo. As both groups had endometriosis in similar proportions at second surgery, these results suggested that interference with estrogen action was related to pain threshold, lowering it in some such that their pain returned sooner.[unreadable] [unreadable] Diagnosis of endometriosis is done at a surgical procedure. One persistent issue in surgical diagnosis is whether histologic confirmation of the disease should be obtained, given the variable appearance of lesions. In the past year, Stratton and Stegmann have correlated biopsy results with lesion appearance in two different ways. In the first study, we reported on the histologic confirmation given varying lesion characteristics, illustrating that no single color was associated with endometriosis and that surgeons should biopsy any suspicious lesion. Overall, it appears that single color lesions had similar frequencies of biopsy-confirmed endometriosis (59 to 62%). Only lesions with multiple colors had a significantly higher percentage of positive biopsies (76%). Of subtle lesions, 60% who only these type of lesions had endometriosis and of these, 40% of women who had only small, subtle lesions had biopsy-proven endometriosis. Mixed color lesions and endometriomas were the only two lesion types that were more commonly biopsy-proven (78%). In a second study, they created a logistic model to predict endometriosis. This model identified characteristics which indicated a high and low probability of biopsy-proven endometriosis. It was useful as a guide in choosing appropriate lesions for biopsy, but should not be used as a substitute for histologic confirmation. [unreadable] [unreadable] Migraine headaches and chronic pelvic pain associated with endometriosis, commonly affect reproductive aged women. We have recently hypothesized that these two chronic, debilitating conditions might co-occur. In our preliminary review of patients enrolled in the clinical trial, at least two thirds of women with chronic pelvic pain have migraine headaches that appear to be independent of endometriosis diagnosis. We will examine whether quality-of-life is lowered, beyond that due to pelvic pain alone. If migraine headache is common in women with chronic pelvic pain, regardless of the presence of endometriosis, it may contribute to disability of those with both conditions and may suggest a common pathophysiology.